Genetic Control of Segmentation in Drosophila: Zygotic Gene Expression
by Dr. William Brook Department of Biochemistry and Molecular Biology, University of Calgary
We previously discussed the maternally
localized factors that control segmentation in the Drosophila embryo. These factors influence the development of large portions of the embryo:
the anterior and posterior halves. During early embryogenesis, these factors become
distributed in concentration gradients. The nuclei in the pre-cellular embryo read these
gradients, resulting in the subdivision of the embryo into domains of gap gene expression.
A generalization is that gap gene expression patterns depend on activation by
transcription factors encoded by maternal and gap genes and is refined by repression by
other gap gene transcription factors ( A good review of this material can be found in
(Hoch and Jäckle, 1993; Kornberg and Tabata, 1993).
How is the maternal gradient information transformed into gap gene expression patterns?
We have looked at hunchback already.
We shall look at Krüppel as another example.
What these genes do together is to define the expression zone of Krüppel
from both the anterior and the posterior sides. Krüppel expression is activated by
Bicoid and low levels of Hunchback throughout most of its region. However, the expression
of Krüppel is repressed on the anterior and posterior sides by high levels of
Hunchback and Knirps, respectively.
Using gel mobility shift assays, they found that both Bicoid and
Knirps bound to this region. Furthermore, using DNase 1 footprinting they found that
Knirps and Bicoid bound to the same 16 bp region within the 730 bp region. They made
plasmid constructs that had seven copies of this 16-mer in front of a CAT gene and
transfected it into Drosophila tissue culture cells
They found that adding a plasmid expressing bicoid
increased the expression of CAT level in a dose-dependent manner. When they subsequently
added a plasmid expressing the knirps gene, they found that they could reduce the
level of CAT expression. This suggests that Bicoid and Knirps regulate the expression of Krüppel
by competing for binding at this 16bp site. When Bicoid binds, Krüppel is
activated. When Knirps binds, Krüppel is repressed.
Bicoid and Hunchback activate even-skipped stripe 2
expression. even-skipped stripe 2 is repressed on the anterior side by the gap gene
giant and on the posterior side by Krüppel.(Small et al., 1991)
Upstream of the even-skipped gene is a 430 bp enhancer
element that controls just the expression of even-skipped in the stripe 2 region
(it is aptly named the "stripe 2 enhancer"!). This enhancer has 12 known factor
binding sites, including 6 activator and 6 repressor sites. The 6 activator sites include
5 Bicoid binding sites and one Hunchback site. There are 3 binding sites for each of Giant
even-skipped regulation is an example of the complex
regulation of spatial gene expression by activation and repression at different control
sites. As an example, in the stripe 1 domain, the stripe 1 enhancer activates even-skipped
transcription and all others are inactive. In the domain between stripe 1 and 2, all seven
enhancers are inactive. In the stripe 2 domain, the stripe 2 enhancer activates
transcription and all others are inactive, etc.
First phase of segment polarity gene expression: pair-rule genes establish segment polarity gene expression patterns
The primary pair-rule gene expression patterns are established by
the coordinate and gap genes and then refined through interactions with each other and
with the secondary pair-rule genes. For example, even-skipped (eve) represses
the expression of fushi-tarazu (ftz)leading to the expression of the two
genes in complementary graded patterns of expression in alternating parasegments.
Other pair-rule genes also control wg and en expression. For example, paired and odd-paired are responsible for the activation of engrailed AND wingless in alternating stripes.
In-class exercise: expression
patterns of wg and en in pair-rule mutants.
Parasegments and segments
These terms can be confusing. Parasegments and segments are
different ways of subdividing the cells along the a/p embryonic axis . They are out of
phase with each other. Parasegments correspond to domains of gene expression and to an
early morphological feature seen before germ-band retraction, the parasegmental groove. At
the time of the cellular blastoderm, the cells within each segment are morphologically
indistinguishable from one another. The boundary between the parasegments is exactly
between the engrailed and wingless expressing cells and is marked during
gastrulation and early germ-band extension by a shallow groove. Beginning after the
completion of germ band extension and during germ band retraction, the parasegmental
grooves disappear at the same time as very deep grooves begin to arise halfway down the
length of each parasegment - these will be the segmental boundaries.
Second phase of segment polarity gene expression: cell to cell signaling
The regulation of the segment polarity genes by the pair-rule
genes is only the first stage of regulation. There are two problems that must be overcome.
First, the expression of the coordinate, gap and pair-rule genes fades away and new
mechanisms for regulating wingless and engrailed are required. Furthermore,
cellularization of the embryo has occurred by this stage and it turns out that the
mechanism for maintaining wingless and engrailed expression is based on cell
to cell communication. This is why not all of the segment-polarity genes are transcription
One of the clues that hedgehog was likely to be part of
the circuit was that it had a phenotype that is very similar to that of wingless. A
series of different genes mutate to the same phenotype (including cubitus interruptus,
gooseberry, smoothened, fused, armadillo, disheveled, porcupine), and most of these
have turned out to be part of either the wingless reception pathway or the hedgehog
(See Gilbert 1997, Figures 14.25 and 14.26)
How do segment polarity genes control pattern in the segment?
The segment polarity gene control the pattern of cell differentiation within each segment. It has been proposed by several groups that wingless and hedgehog form concentration gradients that act as morphogens specifying different fates within the segment much in the same way that bicoid specifies fates within the anterior half of the embryo. For example, removing wingless function with a temperature-sensitive wingless allele after the expression of wingless and engrailed become expressed independently of one another results in a loss of the naked cuticle in each segment. Ubiquitous expression of Wingless in that same phase using a Heat shock-wingless transgene results in the transformation of the denticle belts into naked cuticle. This suggested that the level of Wingless is both necessary and sufficient for the fate of the cells in the naked cuticle.
LEGEND: Wingless and Engrailed expression in the
embryo. The photograph shows a wild-type Drosophila cuticle with the approximate
location of Wingless (blue) and Engrailed (yellow) expressing cells. The two diagrams show
the realtionship between Wingless expressing cells, Engrailed expressing cells, and the
differentiation of cuticle markers in normal embryos (middle) and embryos compromised for
Wingless activity (bottom)
Lawrence and collaborators (1996) examined the effects of manipulating the concentration of Wingless. Embryos that were mutant for both wingless and engrailed lack virtually any evidence of segmentation. By adding back different levels of Wingless using a transgene capable of producing different levels of wingless mRNA, they were able to demonstrate a concentration dependence for Wingless in specifying the fate of cells in the segment. For example, by adding back high levels of Wingless, they were able to produce segments consisting of cuticular elements normally found near the source of Wingless in the segment. Progressively lower levels produced cuticular structures normally found further from the wingless stripe in each segment. (See DiNardo et al., 1994 for a review of the regulation of segment polarity genes.)
Heemskerk and Dinardo (1994) demonstrated that Hedgehog had a concentration dependent effect on the fate of cells in the segmented dorsal epidermis. They accomplished this by either removing Hedgehog activity using a temperature sensitive mutant or by increasing the levels of Hedgehog using a transgene that used the inducible heat shock promoter driving hedgehog transcription.
See Peifer and Bejsovec, 1992, and DiNardo et al., 1994) for reviews of the regulation of segment polarity genes.
Reviews* and References
Arnosti, D. N., Barolo, S., Levine, M., and Small, S. (1996). The
eve stripe 2 enhancer employs multiple modes of transcriptional synergy. Development 122,
Heemskerk, J., and DiNardo, S. (1994). Drosophila hedgehog acts as a morphogen in
cellular patterning. Cell 76, 449-60.
|This material may be reproduced for educational purposes
only provided credit is given to the original source.
Leon Browder & Laurie Iten (Ed.) Dynamic Development
Last revised Monday, November 16, 1998